Anxiety disorders are the most prevalent and costly form of mental illness. Yet, their heterogeneity and high rates of comorbidity have impeded our understanding of underlying pathological mechanisms. The attentional threat bias (AB), widely defined as exaggerated attention towards threat, has been identified as a key cognitive mechanism in anxiety, and posited to be a contributing causal factor. Recent evidence, however, reveals significant heterogeneity of AB and a substantial number of anxious youth with AB away from threat. The clinical importance of understanding AB towards and away from threat as distinct groups with unique neural bases, is highlighted by emerging work showing that AB towards and away from threat predict differential treatment responsiveness and symptom profiles. Currently, we do not fully understand the neurobehavioral profiles underlying AB heterogeneity. Such information will be essential to developing new ways of classifying disorders based on dimensions of observable behavior and brain functions that have the potential to inform targeted interventions. This proposed research directly addresses these questions by testing an empirically- driven neurobehavioral model of AB heterogeneity in adolescents via multimodal neural metrics combined with precise attention assays, and assessing the clinical relevance of these AB neurobehavioral profiles by examining their association with distinct anxiety symptoms. We target early adolescence given the peak in anxiety disorders during these years and significant changes in relevant corticolimbic circuitry. We will combine measurement of intrinsic functional connectivity (iFC) of target corticolimbic and frontoparietal circuits to assess these circuits on the trait level, with temporally-sensitive scalp-recorded event-related potentials (ERPs) and functioning magnetic resonance imaging (fMRI) to assess direct neurocognitive responses to threat stimuli within a large sample of adolescents across the broad spectrum of anxiety severity and diagnostic categories. Aim 1 will be to test the proposed neurobehavioral model by examining whether adolescents with AB towards versus AB away from threat differ on iFC of target circuitry and discrete neural (ERP and fMRI) responses to threat, and by assessing the relationship among these neural and behavioral measures. Aim 2 will be to test the clinical relevance of AB neural and behavioral measures, and the hypothesized neurobehavioral profiles by examining their association with distinct anxiety symptoms ? fear- versus distress-related. Using an innovative combination of brain-based and behavioral methods, this proposal aims to elucidate core subtypes of AB in anxious adolescents. Confirmation of hypotheses would represent a crucial step towards clarifying the conceptualization and measurement of AB and, ultimately, speed progress towards biologically-informed classification and more precise targeted treatment approaches based on AB and its associated neurobiology.